DOCSLIB.ORG

Taylor Expansions and (Log)Linearizing

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Taylor Expansions and (Log)Linearizing Taylor Expansions and (log)linearizing Stéphane Dupraz 1 Mean Value Thoerem Theorem 1.1. Mean Value Theorem in R Let f :[a, b] → R. If f is continuous on [a, b] and differentiable on (a, b), then there exists z ∈ (a, b) such that: f(b) − f(a) = f 0(z)(b − a) Proof. The proof is in two steps. First we prove the result—called Rolle’s theorem—in the particular case when f(b) = f(a). In this case we want to find z such that f 0(z) = 0: a critical point. Since f is continuous on [a, b] compact, f has a maximum and a minimum on [a, b] by Weierstrass theorem. To conclude, we just need to make sure that there is one maximum or minimum on the interior (a, b). There are three cases. First, if f is constant over [a, b] then f 0 = 0 and any z will do. Second, if there exists some x such that f(x) > f(a), then there is an interior maximum. Finally, if there exists some x such that f(x) < f(a), then there is an interior minimum. For the general case, the trick is to define the function g(t) = (f(b) − f(a))t − (b − a)f(t). If satisfies g(b) = g(a), so we can apply Rolle’s theorem to g: there exists z ∈ (a, b) such that g0(z) = f(b)−f(a)−(b−a)f 0(z) = 0. QED. The mean value theorem has a straightforward extension to function from Rn to R. Theorem 1.2. Mean Value Theorem in Rn Let f : S → R, S an open subset of Rn. If f is differentiable, then for any x, y ∈ S such that [x, y] ∈ S, there exists z ∈ (x, y) (that is z = 1 λx + (1 − λ)y for some λ ∈ (0, 1)) such that: f(x) − f(y) = f 0(z)(x − y) Proof. Just define g(λ) = f(λx + (1 − λ)y) for λ ∈ [0, 1] and apply the mean value theorem for function from R to R. 2 2 Taylor Expansions We start with kth-order Taylor expansions for functions from R to R, then consider first and second-order Taylor expansions for functions from Rn to R. 2.1 Taylor formula for functions from R to R We have seen that we can look at the derivative of a function f at a point x as providing the best affine approximation of f around x. We say that we linearize the function f around x. A Taylor expansion of order k of f at x generalizes this notion by looking at the best linear approximation of f around x by a polynomial of order k. There exist variants of the Taylor theorem, depending on the precise meaning of “best approximation” (some require f to be k + 1 times differentiable, other only k times differentiable). Here we state the Taylor-Lagrange theorem, whose proof relies on the mean value theorem. Theorem 2.1. Taylor-Lagrange theorem for functions from R to R Let f :[a, b] → R. If f is k + 1 times differentiable on [a, b], then for all x, y ∈ [a, b], there exists z ∈ (x, y) such that: k X f (i)(x) f (k+1)(z) f(y) = f(x) + (y − x)i + (y − x)k+1 i! (k + 1)! i=1 k X f (i)(x) We call the term f(x) + (y − x)i the Taylor expansion of order k of f at x. i! i=1 We often note the result less rigorously as: k X f (i)(x) f(y) ' f(x) + (y − x)i i! i=1 Proof. The proof is by induction on k. The base case k = 0 is the mean value theorem. The induction step relies on the mean-value theorem. We do not go into the details here. 2.2 Second-order Taylor approximations for functions from Rn to R For functions of n variables, we restrict to first and second-order Taylor expansions. We have already seen first-order Taylor expansions—differentiation/linearization. Stated not too rigorously, the second-order Taylor 3 expansion of a twice differentiable function from Rn to R is 1 f(y) ' f(x) + f 0(x)(y − x) + (y − x)0f 00(x)(y − x) 2 n n n X 1 X X ' f(x) + f 0(x)(y − x ) + f 00 (x)(y − x )(y − x ) i i i 2 ij i i j j j=1 i=1 j=1 The main application of first and second-order Taylor approximations you will see will be in solving Dynamic Stochastic General Equilibrium (DSGE) models in macroeconomics. The perturbations method solves numerically for the equilibrium of an economy defined by a set of difference equations by first taking a first-order or second-order approximation to the equations around the steady-state of the system. 4 3 Linearizing We now take a more applied turn on first-order approximations: linearization. 3.1 Linearizing a function As we have seen, to get the first-order Taylor expansion of a function f at a point x∗—to linearize the function f at x∗—we simply need to calculate its derivative at x∗, f 0(x∗). Then the approximation is f(x) = f(x∗) + f 0(x∗)(x − x∗). Defining dx ≡ x − x∗ and df(x) ≡ f(x) − f(x∗), we can note this as: 0 ∗ 0 ∗ 0 ∗ 0 ∗ df(x) = f (x )dx = f1(x )dx1 + f2(x )dx2 + ... + f (x )dxn and see it as relating the deviation of f(x) from f(x∗) to the deviation of x from x∗. This is most useful to linearize equations. Consider an equation of the form: f(x1, ..., xn) = 0 We can differentiate both sides at x∗ to get: 0 ∗ 0 ∗ 0 ∗ 0 ∗ f (x )dx = f1(x )dx1 + f2(x )dx2 + ... + fn(x )dxn = 0 This turns the equation that relates the variables x1,...,xn into an linear equation that relates the deviations of each variables dx1, ..., dxn. 3.2 Practical rules The rules of differentiation can be easily expressed in those terms. The linearity of the differentiation operator and the chain rule translate into: 1. d(x + y) = dx + dy. 2. d(λx) = λdx. 3. df(g(x)) = f 0(g(x∗))dg(x) = f 0(g(x∗))g0(x∗)dx. For example, to linearize a resource constraint: f(Kt) = Ct + Kt+1 − (1 − δ)Kt 5 ∗ ∗ around Ct = C and Kt+1 = Kt = K : 0 ∗ f (K )dKt = dCt + dKt+1 − (1 − δ)dKt 6 4 Loglinearizing In economics, it is often more relevant to focus on relative deviations (or percentage deviations) rather than on absolute deviations (or level deviations). For instance, knowing that “consumption is 5% below its steady-state value” is more meaningful than knowing that “consumption is $100 below its steady-state value”, the meaning of which depends on how high steady-state consumption is. We are more interested in the relative deviation variables, that we will denote with a hat: dx x − x∗ x = = b x∗ x∗ But it turns out that at first order1: dx x = = dx × ln0(x∗) = d ln(x) b x∗ where the last term means (as for any function): d ln(x) = ln(x)−ln(x∗). For this reason, turning a function—a relationship between inputs x1, ..., xn and output f(x)—into a linear function of the relative variation in x, xˆ1,...,xˆn, and the relative variation in y, yˆ, is called loglinearizing the function. (Turning an equation— a relationship between variables x1,...,xn—into a linear equation that relates the relative variations of each variables xˆ1, ..., xˆn, is called loglinearizing the equation). 4.1 Formally Formally, to loglinearize a function f(x) instead of linearizing it, we first do some changes of variables. • Instead of the variation dxi, we want the variation d ln(xi), so we define yi = ln(xi). • Instead of the variation df(x), we want the variation d ln(f(x)), so we consider the function ln(f(x)). This way, we consider the function: g(y) = ln(f(ey1 , ..., eyn )) ∗ ∗ 1You can also see it through : dln(x) = ln(x) − ln(x∗) = ln x = ln x−x + 1 ' x−x ≡ x. x∗ x∗ x∗ b 7 Linearizing: ∗ n 0 y∗,...,eyn f (e 1 ) ∗ X i yi dg(y) = y∗ y∗ e dyi f(e 1 , ..., e n ) i=1 n X f 0(x∗) d ln(f(x)) = i x∗d ln(x ) f(x∗) i i i=1 ∗ Definition 4.1. The elasticity of a function f at x with respect to xi is: 0 ∗ ∗ i ∗ fi (x ) ∗ ∂ ln(f(x )) εf (x ) = ∗ xi = f(x ) ∂ ln(xi) ∗ It expresses the percentage change of f(x) for a percentage increase in xi of 1% (at x ). We finally get: ∗ ∗ fd(x) = ε1(x )ˆx1 + ... + εn(x )ˆxn In practice, we can rely on two approaches to loglinearize a function or an equation. 4.2 First Approach and Practical Rules In practice, we can use the following method: Approach 1. To loglinearize a function f: 1. Take the log of f. 2. Linearize ln(f), remembering dln(x) = xb. For instance let us loglinearize a Cobb-Douglas production function Y = AKαL1−α at any point: ln(Y ) = ln(A) + α ln(K) + (1 − α) ln(L) d ln(Y ) = d ln(A) + αd ln(K) + (1 − α)d ln(L) Yˆ = Aˆ + αKˆ + (1 − α)Lˆ This example is straightforward because of the following practical rules for loglinearization, which are simply the translation of the corresponding rules for linearization: 8 1.
Load more

Recommended publications
  • 1 Probelms on Implicit Differentiation 2 Problems on Local Linearization
    Math-124 Calculus Chapter 3 Review 1 Probelms on Implicit Di®erentiation #1. A function is de¯ned implicitly by x3y ¡ 3xy3 = 3x + 4y + 5: Find y0 in terms of x and y. In problems 2-6, ¯nd the equation of the tangent line to the curve at the given point. x3 + 1 #2. + 2y2 = 1 ¡ 2x + 4y at the point (2; ¡1). y 1 #3. 4ey + 3x = + (y + 1)2 + 5x at the point (1; 0). x #4. (3x ¡ 2y)2 + x3 = y3 ¡ 2x ¡ 4 at the point (1; 2). p #5. xy + x3 = y3=2 ¡ y ¡ x at the point (1; 4). 2 #6. x sin(y ¡ 3) + 2y = 4x3 + at the point (1; 3). x #7. Find y00 for the curve xy + 2y3 = x3 ¡ 22y at the point (3; 1): #8. Find the points at which the curve x3y3 = x + y has a horizontal tangent. 2 Problems on Local Linearization #1. Let f(x) = (x + 2)ex. Find the value of f(0). Use this to approximate f(¡:2). #2. f(2) = 4 and f 0(2) = 7. Use linear approximation to approximate f(2:03). 6x4 #3. f(1) = 9 and f 0(x) = : Use a linear approximation to approximate f(1:02). x2 + 1 #4. A linear approximation is used to approximate y = f(x) at the point (3; 1). When ¢x = :06 and ¢y = :72. Find the equation of the tangent line. 3 Problems on Absolute Maxima and Minima 1 #1. For the function f(x) = x3 ¡ x2 ¡ 8x + 1, ¯nd the x-coordinates of the absolute max and 3 absolute min on the interval ² a) ¡3 · x · 5 ² b) 0 · x · 5 3 #2.
    [Show full text]
  • DYNAMICAL SYSTEMS Contents 1. Introduction 1 2. Linear Systems 5 3
    DYNAMICAL SYSTEMS WILLY HU Contents 1. Introduction 1 2. Linear Systems 5 3. Non-linear systems in the plane 8 3.1. The Linearization Theorem 11 3.2. Stability 11 4. Applications 13 4.1. A Model of Animal Conflict 13 4.2. Bifurcations 14 Acknowledgments 15 References 15 Abstract. This paper seeks to establish the foundation for examining dy- namical systems. Dynamical systems are, very broadly, systems that can be modelled by systems of differential equations. In this paper, we will see how to examine the qualitative structure of a system of differential equations and how to model it geometrically, and what information can be gained from such an analysis. We will see what it means for focal points to be stable and unstable, and how we can apply this to examining population growth and evolution, bifurcations, and other applications. 1. Introduction This paper is based on Arrowsmith and Place's book, Dynamical Systems.I have included corresponding references for propositions, theorems, and definitions. The images included in this paper are also from their book. Definition 1.1. (Arrowsmith and Place 1.1.1) Let X(t; x) be a real-valued function of the real variables t and x, with domain D ⊆ R2. A function x(t), with t in some open interval I ⊆ R, which satisfies dx (1.2) x0(t) = = X(t; x(t)) dt is said to be a solution satisfying x0. In other words, x(t) is only a solution if (t; x(t)) ⊆ D for each t 2 I. We take I to be the largest interval for which x(t) satisfies (1.2).
    [Show full text]
  • Linearization of Nonlinear Differential Equation by Taylor's Series
    International Journal of Theoretical and Applied Science 4(1): 36-38(2011) ISSN No. (Print) : 0975-1718 International Journal of Theoretical & Applied Sciences, 1(1): 25-31(2009) ISSN No. (Online) : 2249-3247 Linearization of Nonlinear Differential Equation by Taylor’s Series Expansion and Use of Jacobian Linearization Process M. Ravi Tailor* and P.H. Bhathawala** *Department of Mathematics, Vidhyadeep Institute of Management and Technology, Anita, Kim, India **S.S. Agrawal Institute of Management and Technology, Navsari, India (Received 11 March, 2012, Accepted 12 May, 2012) ABSTRACT : In this paper, we show how to perform linearization of systems described by nonlinear differential equations. The procedure introduced is based on the Taylor's series expansion and on knowledge of Jacobian linearization process. We develop linear differential equation by a specific point, called an equilibrium point. Keywords : Nonlinear differential equation, Equilibrium Points, Jacobian Linearization, Taylor's Series Expansion. I. INTRODUCTION δx = a δ x In order to linearize general nonlinear systems, we will This linear model is valid only near the equilibrium point. use the Taylor Series expansion of functions. Consider a function f(x) of a single variable x, and suppose that x is a II. EQUILIBRIUM POINTS point such that f( x ) = 0. In this case, the point x is called Consider a nonlinear differential equation an equilibrium point of the system x = f( x ), since we have x( t )= f [ x ( t ), u ( t )] ... (1) x = 0 when x= x (i.e., the system reaches an equilibrium n m n at x ). Recall that the Taylor Series expansion of f(x) around where f:.
    [Show full text]
  • Linearization Extreme Values
    Math 31A Discussion Notes Week 6 November 3 and 5, 2015 This week we'll review two of last week's lecture topics in preparation for the quiz. Linearization One immediate use we have for derivatives is local linear approximation. On small neighborhoods around a point, a differentiable function behaves linearly. That is, if we zoom in enough on a point on a curve, the curve will eventually look like a straight line. We can use this fact to approximate functions by their tangent lines. You've seen all of this in lecture, so we'll jump straight to the formula for local linear approximation. If f is differentiable at x = a and x is \close" to a, then f(x) ≈ L(x) = f(a) + f 0(a)(x − a): Example. Use local linear approximation to estimate the value of sin(47◦). (Solution) We know that f(x) := sin(x) is differentiable everywhere, and we know the value of sin(45◦). Since 47◦ is reasonably close to 45◦, this problem is ripe for local linear approximation. We know that f 0(x) = π cos(x), so f 0(45◦) = πp . Then 180 180 2 1 π 90 + π sin(47◦) ≈ sin(45◦) + f 0(45◦)(47 − 45) = p + 2 p = p ≈ 0:7318: 2 180 2 90 2 For comparison, Google says that sin(47◦) = 0:7314, so our estimate is pretty good. Considering the fact that most folks now have (extremely powerful) calculators in their pockets, the above example is a very inefficient way to compute sin(47◦). Local linear approximation is no longer especially useful for estimating particular values of functions, but it can still be a very useful tool.
    [Show full text]
  • Linearization Via the Lie Derivative ∗
    Electron. J. Diff. Eqns., Monograph 02, 2000 http://ejde.math.swt.edu or http://ejde.math.unt.edu ftp ejde.math.swt.edu or ejde.math.unt.edu (login: ftp) Linearization via the Lie Derivative ∗ Carmen Chicone & Richard Swanson Abstract The standard proof of the Grobman–Hartman linearization theorem for a flow at a hyperbolic rest point proceeds by first establishing the analogous result for hyperbolic fixed points of local diffeomorphisms. In this exposition we present a simple direct proof that avoids the discrete case altogether. We give new proofs for Hartman’s smoothness results: A 2 flow is 1 linearizable at a hyperbolic sink, and a 2 flow in the C C C plane is 1 linearizable at a hyperbolic rest point. Also, we formulate C and prove some new results on smooth linearization for special classes of quasi-linear vector fields where either the nonlinear part is restricted or additional conditions on the spectrum of the linear part (not related to resonance conditions) are imposed. Contents 1 Introduction 2 2 Continuous Conjugacy 4 3 Smooth Conjugacy 7 3.1 Hyperbolic Sinks . 10 3.1.1 Smooth Linearization on the Line . 32 3.2 Hyperbolic Saddles . 34 4 Linearization of Special Vector Fields 45 4.1 Special Vector Fields . 46 4.2 Saddles . 50 4.3 Infinitesimal Conjugacy and Fiber Contractions . 50 4.4 Sources and Sinks . 51 ∗Mathematics Subject Classifications: 34-02, 34C20, 37D05, 37G10. Key words: Smooth linearization, Lie derivative, Hartman, Grobman, hyperbolic rest point, fiber contraction, Dorroh smoothing. c 2000 Southwest Texas State University. Submitted November 14, 2000.
    [Show full text]
  • Linearization and Stability Analysis of Nonlinear Problems
    Rose-Hulman Undergraduate Mathematics Journal Volume 16 Issue 2 Article 5 Linearization and Stability Analysis of Nonlinear Problems Robert Morgan Wayne State University Follow this and additional works at: https://scholar.rose-hulman.edu/rhumj Recommended Citation Morgan, Robert (2015) "Linearization and Stability Analysis of Nonlinear Problems," Rose-Hulman Undergraduate Mathematics Journal: Vol. 16 : Iss. 2 , Article 5. Available at: https://scholar.rose-hulman.edu/rhumj/vol16/iss2/5 Rose- Hulman Undergraduate Mathematics Journal Linearization and Stability Analysis of Nonlinear Problems Robert Morgana Volume 16, No. 2, Fall 2015 Sponsored by Rose-Hulman Institute of Technology Department of Mathematics Terre Haute, IN 47803 Email: [email protected] a http://www.rose-hulman.edu/mathjournal Wayne State University, Detroit, MI Rose-Hulman Undergraduate Mathematics Journal Volume 16, No. 2, Fall 2015 Linearization and Stability Analysis of Nonlinear Problems Robert Morgan Abstract. The focus of this paper is on the use of linearization techniques and lin- ear differential equation theory to analyze nonlinear differential equations. Often, mathematical models of real-world phenomena are formulated in terms of systems of nonlinear differential equations, which can be difficult to solve explicitly. To overcome this barrier, we take a qualitative approach to the analysis of solutions to nonlinear systems by making phase portraits and using stability analysis. We demonstrate these techniques in the analysis of two systems of nonlinear differential equations. Both of these models are originally motivated by population models in biology when solutions are required to be non-negative, but the ODEs can be un- derstood outside of this traditional scope of population models.
    [Show full text]
  • 1 Introduction 2 Linearization
    I.2 Quadratic Eigenvalue Problems 1 Introduction The quadratic eigenvalue problem (QEP) is to find scalars λ and nonzero vectors u satisfying Q(λ)x = 0, (1.1) where Q(λ)= λ2M + λD + K, M, D and K are given n × n matrices. Sometimes, we are also interested in finding the left eigenvectors y: yH Q(λ) = 0. Note that Q(λ) has 2n eigenvalues λ. They are the roots of det[Q(λ)] = 0. 2 Linearization A common way to solve the QEP is to first linearize it to a linear eigenvalue problem. For example, let λu z = , u Then the QEP (1.1) is equivalent to the generalized eigenvalue problem Lc(λ)z = 0 (2.2) where M 0 D K L (λ)= λ + ≡ λG + C. c 0 I −I 0 Lc(λ) is called a companion form or a linearization of Q(λ). Definition 2.1. A matrix pencil L(λ)= λG + C is called a linearization of Q(λ) if Q(λ) 0 E(λ)L(λ)F (λ)= (2.3) 0 I for some unimodular matrices E(λ) and F (λ).1 For the pencil Lc(λ) in (2.2), the identity (2.3) holds with I λM + D λI I E(λ)= , F (λ)= . 0 −I I 0 There are various ways to linearize a quadratic eigenvalue problem. Some are preferred than others. For example if M, D and K are symmetric and K is nonsingular, then we can preserve the symmetry property and use the following linearization: M 0 D K L (λ)= λ + .
    [Show full text]
  • Graduate Macro Theory II: Notes on Log-Linearization
    Graduate Macro Theory II: Notes on Log-Linearization Eric Sims University of Notre Dame Spring 2017 The solutions to many discrete time dynamic economic problems take the form of a system of non-linear difference equations. There generally exists no closed-form solution for such problems. As such, we must result to numerical and/or approximation techniques. One particularly easy and very common approximation technique is that of log linearization. We first take natural logs of the system of non-linear difference equations. We then linearize the logged difference equations about a particular point (usually a steady state), and simplify until we have a system of linear difference equations where the variables of interest are percentage deviations about a point (again, usually a steady state). Linearization is nice because we know how to work with linear difference equations. Putting things in percentage terms (that's the \log" part) is nice because it provides natural interpretations of the units (i.e. everything is in percentage terms). First consider some arbitrary univariate function, f(x). Taylor's theorem tells us that this can be expressed as a power series about a particular point x∗, where x∗ belongs to the set of possible x values: f 0(x∗) f 00 (x∗) f (3)(x∗) f(x) = f(x∗) + (x − x∗) + (x − x∗)2 + (x − x∗)3 + ::: 1! 2! 3! Here f 0(x∗) is the first derivative of f with respect to x evaluated at the point x∗, f 00 (x∗) is the second derivative evaluated at the same point, f (3) is the third derivative, and so on.
    [Show full text]
  • Chapter 3. Linearization and Gradient Equation Fx(X, Y) = Fxx(X, Y)
    Oliver Knill, Harvard Summer School, 2010 An equation for an unknown function f(x, y) which involves partial derivatives with respect to at least two variables is called a partial differential equation. If only the derivative with respect to one variable appears, it is called an ordinary differential equation. Examples of partial differential equations are the wave equation fxx(x, y) = fyy(x, y) and the heat Chapter 3. Linearization and Gradient equation fx(x, y) = fxx(x, y). An other example is the Laplace equation fxx + fyy = 0 or the advection equation ft = fx. Paul Dirac once said: ”A great deal of my work is just playing with equations and see- Section 3.1: Partial derivatives and partial differential equations ing what they give. I don’t suppose that applies so much to other physicists; I think it’s a peculiarity of myself that I like to play about with equations, just looking for beautiful ∂ mathematical relations If f(x, y) is a function of two variables, then ∂x f(x, y) is defined as the derivative of the function which maybe don’t have any physical meaning at all. Sometimes g(x) = f(x, y), where y is considered a constant. It is called partial derivative of f with they do.” Dirac discovered a PDE describing the electron which is consistent both with quan- respect to x. The partial derivative with respect to y is defined similarly. tum theory and special relativity. This won him the Nobel Prize in 1933. Dirac’s equation could have two solutions, one for an electron with positive energy, and one for an electron with ∂ antiparticle One also uses the short hand notation fx(x, y)= ∂x f(x, y).
    [Show full text]
  • Note: Linearization @ 2020-01-21 01:01:05-08:00
    Note: Linearization @ 2020-01-21 01:01:05-08:00 EECS 16B Designing Information Devices and Systems II Fall 2019 Note: Linearization Overview So far, we have spent a great deal of time studying linear systems described by differential equations of the form: d ~x(t) = A~x(t) + B~u(t) +~w(t) dt or linear difference equations (aka linear recurrence relations) of the form: ~x(t + 1) = A~x(t) + B~u(t) +~w(t): Here, the linear difference equations came from either discretizing a continuous model or were learned from data. In the real world, however, it is very rare for physical systems to perfectly obey such simple linear differential equations. The real world is generally continuous, but our systems will obey nonlinear differential equations of the form: d ~x = ~f (~x(t);~u(t)) +~w(t); dt where ~f is some non-linear function. Through linearization, we will see how to approximate general systems as locally linear systems, which will in turn allow us to apply all the techniques we have developed so far. 1 Taylor Expansion First, we must recall how to approximate the simplest nonlinear functions — scalar functions of one variable — as linear functions. To do so, we will use the Taylor expansion we learned in calculus. This expansion tells us that for well-behaved functions f (x), we can write 2 d f 1 d f 2 f (x) = f (x0) + (x − x0) + (x − x0) + ··· dx 2 dx2 x=x0 x=x0 near any particular point x = x0, known as the expansion point.
    [Show full text]
  • 32-Linearization.Pdf
    Math S21a: Multivariable calculus Oliver Knill, Summer 2011 1 What is the linear approximation of the function f(x, y) = sin(πxy2) at the point (1, 1)? We 2 2 2 have (fx(x, y),yf (x, y)=(πy cos(πxy ), 2yπ cos(πxy )) which is at the point (1, 1) equal to f(1, 1) = π cos(π), 2π cos(π) = π, 2π . Lecture 10: Linearization ∇ h i h− i 2 Linearization can be used to estimate functions near a point. In the previous example, 0.00943 = f(1+0.01, 1+0.01) L(1+0.01, 1+0.01) = π0.01 2π0.01+3π = 0.00942 . In single variable calculus, you have seen the following definition: − ∼ − − − The linear approximation of f(x) at a point a is the linear function 3 Here is an example in three dimensions: find the linear approximation to f(x, y, z)= xy + L(x)= f(a)+ f ′(a)(x a) . yz + zx at the point (1, 1, 1). Since f(1, 1, 1) = 3, and f(x, y, z)=(y + z, x + z,y + ∇ − x), f(1, 1, 1) = (2, 2, 2). we have L(x, y, z) = f(1, 1, 1)+ (2, 2, 2) (x 1,y 1, z 1) = ∇ · − − − 3+2(x 1)+2(y 1)+2(z 1)=2x +2y +2z 3. − − − − 4 Estimate f(0.01, 24.8, 1.02) for f(x, y, z)= ex√yz. Solution: take (x0,y0, z0)=(0, 25, 1), where f(x0,y0, z0) = 5. The gradient is f(x, y, z)= y=LHxL x x x ∇ (e √yz,e z/(2√y), e √y).
    [Show full text]
  • 1.4 Stability and Linearization
    42 Introduction 1.4 Stability and Linearization Suppose we are studying a physical system whose states x are described by points in Rn. Assume that the dynamics of the system is governed by a given evolution equation dx = f(x). dt Let x0 be a stationary point of the dynamics, i.e., f(x0) = 0. Imagine that we perform an experiment on the system at time t = 0 and determine that the initial state is indeed x0. Are we justified in predicting that the system will remain at x0 for all future time? The mathematical answer to this question is yes, but unfortunately it is probably not the question we really wished to ask. Experiments in real life seldom yield exact answers to our idealized models, so in most cases we will have to ask whether the system will remain near x0 if it started near x0. The answer to the revised question is not always yes, but even so, by examining the evolution equation at hand more carefully, one can sometimes make predictions about the future behavior of a system starting near x0. A simple example will illustrate some of the problems involved. Consider the following two differential equations on the real line: x!(t)= x(t) (1.4.1) − and x!(t)=x(t). (1.4.2) The solutions are, respectively, t x(x0,t)=x0e− (1.4.3) and +t x(x0,t)=x0e . (1.4.4) Note that 0 is a stationary point of both equations. In the first case, for all x0 R, we have limt x(x0,t) = 0.
    [Show full text]